DOCUMENT RESUME

ED 298 122 SP 030 554

AUTHOR Berliner, David C.TITLE The Development of Expertise in Pedagogy.INSTITUTION American Association of Colleges for Teacher

Education, Washington, D.C.REPORT NO ISBN-0-89333-053-1PUB DATE Feb 88NOTE 35p.; Charles W. Hunt Memorial Lecture presented at

the Annual Meeting of the American Association ofColleges for Teacher Education (New Orleans, LA,February 17-20, 1988).

AVAILABLE FROM AACTE Publications, One Dupont Circle, Suite 610,Washington, DC 20036-2412 ($12.00).

PUB TYPE Viewpoints (120) -- Reports - Descriptive (141) --Speeches /Conference Papers (150)

EDRS PRICE MF01/PCO2 Plus Postage.DESCRIPTORS *Beginning Teachers; *Developmental Stages;

Elementary Secondary Education; Higher Education;Policy Formation; Preservice Teacher Education;*Skill Development; *Teacher Characteristics;*Teacher Education Programs; Teacher Effectiveness;Teacher Evaluation

ABSTRACTIn this discussion on the development of expertise in

teaching, a theory of skill learning is first presented. Thecharacteristics of five stages of skill development in teachers aredescribed: (1) novice; (2) advanced beginner; (3) competent teacher;(4) proficient teacher; and (5) expert teacher. A review of datacollected by studies on the subject of teaching expertise points outdifferences between the novice and the expert teacher in the areasof: (1) interpreting classroom phenomena; (2) discerning theimportance of events; (3) using routines; (4) predicting classroomphenomena; (5) judging typical and atypical events; and (6)evaluating performance: responsibility and emotions. The discussionof policy considerations for teacher educators, based upon thisdevelopmental theory of skill acquisition, is aimed at helpingnovices become proficient in classroom techniques while evaluatingthem in ways approriate for their developmental level. (JD)

XXXXXXX30000(***********KM************MENKMMXXXXXXXXXXXX****************

Reproductions supplied by EDRS are the best that can be madefrom the original document.

300000000a****Y****X*******************K*******************************

THE DEVELOPMENT OFEXPERTISE IN PEDAGOGY

David C. Berliner

Professor of Psychology in Education and Curriculum and instruction,Arizona State University

and

Spencer Fellow, Center for Advanced Studyin the Behavioral Sciences

Charles W. Hunt Memorial LectureAmerican Association of Colleges for Teacher Education

New Orleans, La., February 17, 1988

3

The Charles W. Hunt Memorial Lecture at AACTE's 1988 Annual Meeting and publi-cation of this document were supported in part by a grant from the people of Chevron U.S.A.Inc.

The opinions, conclusions, and recommendations expressed in this report are those ofthe author; they do not necessarily reflect those of the American Association of Colleges forTeacher Education or of Chevron U.S.A. Inc. AACTE is printing and distributing this docu-ment to stimulate discussion, study, and improvement of teacher education.

Copies of The Development of Expertise in Pedagogy may be ordered from:

AACTE PublicationsOne Dupont Circle, Suite 610Washington, DC 20036-2412

Single copy $12.00 prepaid.

Copyright 0 1988 by the American Association of Colleges for Teacher Education.

All rights reserved.

Printed in the United States of America.

Library of Congress Catalog Number 88-071816

International Standard Book Number 0-89333-053-1

CONTENTS

Charles W. Hunt iv

A Theory of Skill Learning 1

Stage 1: Novice 2

Stage 2: Advanced Beginner 3

Stage 3: Competent 4

Stage 4: Proficient 4

Stage 5: Expert 5

Summary 6

Studies of Expertise 7

Interpreting classroom phenomena 7

Discerning the importance of events 12

Using routines 13

Predicting classroom phenomena 15

Judging typical and atypical events 16

Evaluating performance: Responsibility and emotions 18

Miscellaneous 19

Policy Considerations 20

Conclusion 27

References 28

iii 5

)

CHARLES W. HUNT1880-1973

With a passion for teaching and a love of people, Charles WesleyHunt helped shape teacher education in America for nearly half a cen-tury. His career spanned the range of educational responsibilitiesteacher, university dean, president of the State Teachers College at One-onta, N.Y., and volunteer in national associations for teacher education.

As secretary-treasurer first of the American Association of TeachersColleges and subsequently the American Association of Colleges forTeacher Education (AACTE), which he helped to create, Dr. Hunt par-ticipated directly in the changes sweeping teacher education during themid-20th century. He worked diligently to develop our national associa-tion as the vehicle to stimulate and effect necessary ch 'es in the educa-tion of teachers. The tools for change were varied, but of special sig-nificance were institutional accreditation, qualitative standards for effec-tive programs, and inclusion of all types of higher education institutions.

When the lecture series honoring him was established in 1960, Dr.Hunt stated:

In the American Association of Colleges for Teacher Education we havecome from our varying stations across the nation to share our exper-ience, to pool our strength, and to play our role in the galaxy of institu-tional organizations which are very important in our national culture.The gradual assembling of all (collegiate) institutions for the preparationof teachers into one working group is a movement of great significance.

Today is the future that Charlie Hunt could only dream about. It isthe future his life's work made possible. While I am sure he would applaudour accomplishments, I am equally sure he would urge us to look beyondour horizon, to anticipate the challenges of the future, and to prepare our-selves to meet that future with understanding and enthusiasm.

6iv

Edward C. PomeroyExecutive Director

emeritus

AACTE

June 1988

THE DEVELOPMENT OFEXPERTISE IN PEDAGOGY

We can learn a good deal from studies about how expert ice skaters,bridge players, radiologists, nurses, mathematicians, and even full -timeracetrack handicappers apply their domain-specific knowledge to par-ticular problems they face. In the last few years experts in such diversefields have been studied and remarkabl : similarities in the ways they per-ceive and process information have been shown to exist (Chi, Glaser, &Farr, in press; Berliner, 1986). But the path one travels from novice to ex-pert in these various fields has not yet been well studied, in part becauseresearch on expertise is relatively new, and in part because longitudinalstudies are needed, and these are among the least likely to receive support.Still, if we had a general theory about the development of expertise in dif-ferent fields and data about the ways experts and novices perform peda-gogical tasks, we might discover that there are policy consietaations of in-terest to the teacher education community. Moreover, some of thesepolicy considerations may not be obvious. Thus, although quite tentative,I would like first to speculate about a general theory of the development ofexpertise. This theory provides a way to think about the stages one movesthrough as ignorance is overcome and expertise is achieved in a particularfield. Then I will present some interpretations of data my colleagues and Ihave collected in studies of pedagogical expertise that inform this theory.Finally, I will try to draw out policy considerations for teacher educators aswe think about the journey from novice teacher to expert.

A Theory of Skill Let us view the development of expertise in pedagogy as consisting ofLearning five stages of skill development, following the general model presented by

two Berkeley professors, the philosopher Hubert Dreyfus and his brother,computer scientist Stuart Dreyfus (1986). We begin with novices, who,with experience, develop into advanced beginners. Most of these indi-viduals then become competent teachers. It should be the goal of teacher

7

`The behavior of

the novice is ra-

tional, relatively

inflexible, and

*Inds to conform to

whatever rules and

procedures they

were told to follow.'

le.gzaEs..=-

education colleges to help prepare the novice and assist the advancedbeginner to become a competent teacher. Competence, I believe, shouldbe our goal. Pursuing more or talking of excellence or expertise for allteachers is to pursue a will-o-the-wisp. And to deny that differences inteaching ability exist, as some professional organizations have done, is toperpetuate a fiction. We all know better. Although individual differencesabound, I would hypothesize that novices are generally student and begin-ning first-year teachers. Advanced beginners are often in the second andthird years of their teaching careers, and if they have an talent andmotivation whatsoever, about the third or fourth year they may becomecompetent. I am convinced that this stage of development can be achievedby the vast majority of graduates of our teacher training cc:leges. Perhapsaround the fifth year a modest number of teachers may me re into a furtherstage of development, that of proficient. Some of these p' licient teacherswill reach the highest stage, achieved by few members ci a field, that of ex-pert. Each of these stages has some distinguishing characteristics, to whichwe now turn.

Stage 1: Novice. Here the commonplaces must be discerned. Theelements of the tasks to be performed need to be labeled and learned, andone learns a set of context-free rules to guide behavior. For example, inlearning to drive an automobile the novice is taught the meaning of high-way signs, lane markings, safety regulations, etiquette at stop signs, andother elements of the task. Novices are taught to shift at about 12 milesper hour, a rule that is virtually context free, to be followed on upgradesand downgrades, in good weather or bad, in slow-moving traffic or onempty streets. In learning to teach, the novice is taught the meaning ofterms like higher-order questions, reinforcement, and learning disabled.Novices are taught context-free rules such as "give praise for .fight an-swers," "wait at least three seconds after asking a higher-order question,""never criticize a student," and that old stand by, "never smile untilChristmas." Identifying the context-free elements and rules is needed inorder to begin to teach. The behavior of the novice is rational, relativelyinflexible, and tends to conform to whatever rules and procedures theywere told to follow. Only marginal performance is really expected. This is

2 8

a stage for learning the objective facts and features of situations. It is astage for gaining experience. And it is the stage at which real-world ex-perience appears to be far more important than verbal information, asgenerations of student teachers have informed us.

Stage 2: Advanced beginner. Here experience can become meldedwith verbal knowledge, similarities across contexts are recognized, andepisodic knowledge is built up. Strategic knowledge when to ignore orbreak rules and when to follow themis developed as context begins toguide behavior. For example, when driving on ice you need the torquegenerated in second gear and so it might be neccessary to stay in that gearlonger. In teaching, you learn that praise does not always have the desiredeffect, such as when a low-ability child interprets it as communicating lowexpectations. And you learn that criticism of a usually good student aftera bad performance can be quite motivating. Experience is affecting be-havior, but the advanced beginner still has no sense of what is important.Benner (1984, p. 23-24) makes this point in comparing novices and ad-vanced beginners with competent nurses.

1 give instructions to the new graduate, very detailed and explicit in-structions: When you come in and first see the baby, you take the baby'svital signs and make the physical examination, and you check the I. V.sites, and the ventilator and make sure that it works, and you check themonitors and alarms. When I would say this to them, they would do ex-actly what I told them to do, no matter what else was going on...Theycouldn't choose one to leave out. They couldn't choose which was themost important...They couldn't do for one baby the things that weremost important and then go on to the other baby and do the things thatwere most important, and leave the things that weren't as importantuntil later on... If I said, you have to do these eight things...they didthose things, and they didn't care if their other kid was screaming itshead off. When they did realize, they would be like a mule between cwopiles of hay.

The novice and the advanced beginner, though intensely involved inthe learning process, may also lack a certain responsibility for their ac-tions. This occurs because they are labeling and describing events, follow-

3 9

`Competent per-

formers . . . have

rational goals and

choose sensible

means for reaching

them [and) they can

determine what is

and what is not

important.'

ing rules, recognizing and classifying contexts, but not actively determin-ing through personal action what is happening. The acceptance of per-sonal responsibility for classroom instruction occurs when personaldecision making, willfully choosing what to do, takes place. This occurs inthe next stage of development.

Stage 3: Competent. Competent performers of a skill have two dis-tinguishing characteristics. First, they make conscious choices about whatthey are going to do. They set pric rities and decide on plans. They haverational goals and choose sensible means for reaching them. In addition,while enacting their skill, they can eetermine what is and what is not im-portant. From their experience they know what to attend to and what toignore. In driving, this is the stage wh !re one can shorten time in going tothe airport by changing lanes, speedinl up, and committing minor infrac-tions of the law, but always under con: cious control and with knowledgeof what is safe and what is not. In teaching, this is the stave where onelearns not to make timing and targetin,: errors, because one has learnedthrough experience what to attend to i nd what to ignore. And this iswhen teachers learn to make curriculum gad instruction decisions, such aswhen to stay with a topic and when to me ve on.

Because they are more personally in control of the events aroundthem, following their own plans, and respc riding only to the informationthat they choose to, teachers at this stage tend to feel more responsibilityfor what happens. They are not detached. Thus, they often feel emotionalabout success and failure in a different and more intense way than donovices or advanced beginners. And they have more vivid memories oftheir successes and failures as well. But competent performers are not yetfast, fluid, or flexible in their behavior. These are characteristics of the lasttwo stages in the development of expertise.

Stage 4: Proficient. This is the stage at which intuition or know-howbecomes prominent. Nothing mysterious is meant by these terms. Thinkof the microadjustments made in riding a bicycle. At some point in learn-ing to ride a bike you no longer think about thew adjustments. You havean "intuitive" sense of the situation. Further, from the wealth of exper-ience that the proficient individual has accumulated comes a holistic

4 10

recognition of similarities. For example, it is the stage when a teachernotices, without conscious effort, that today's mathematics lesson is bog-ging down for the same reason that last week's spelling lesson bombed. Atsome higher level of categorization the similarities between disparateevents is understood. This holistic similarity recognition allows proficientindividuals to predict events more precisely because they see more thingsas alike and, therefore, as having been experienced before. Chess masters,bridge masters, air traffic controllers, and radiologists rely upon this abil-ity. The proficient performer, however, while intuitive in pattern recogni-tion and in ways of knowing, is still analytic and deliberative in decidingwhat to do. An example of this distinction may be found by considering aproficient automobile driver approaching a curve on a rainy day. It is byintuition developed through accumulated experience that the driver sim-ply "knows" when the car is going too fast. An analysis of options then oc-curs and the driver decides whether to brake, let up on the gas, or downshift. Most tournament chess and bridge players are at this proficientstage. But the grand masters are those few who move to a stage higher, tothe expert level.

Stage 5: Expert. If novices, advanced beginners, and competent per-formers are rational, and proticient performers are intuitive, we mightcategorize experts as "arational." They have an intuitive grasp of a situa-tion and seem to sense in nonanalytic, nondeliberative ways the ap-propriate response to make. They show fluid performance, as we all dowhen we no longer have to choose our words when speaking . r thinkabout where to place our feet when walking. We simply talk and walk inan apparently effortless manner. The expert safety in football, the expertmartial artist in combat, the expert chess master, and the expert teacherin classroom recitations all seem to know where to be or what to do at theright time. They engage in their performance in a qualitatively differentway than do novices or competent performers. The pilots who no longerfly the plane but believe they are personally flying, the race car driverswho talk of "becoming one with the machine," and the air traffic control-lers who see airplanes rather than blips on the screen are involved in theirtasks in qualitatively different ways than are others. They are not con-

511

[1.11WMTIMkilAgiliaMililli.ilail.1211 _-_

` . . . when things

are going smoothly,

experts rarely

appear to be

reflective about

their performance.'

sciously choosing what to attend to and what to do. They are acting ef-fortlessly and fluidly, behaving in ways that are not easily described asdeductive or analytic. Though beyond the usual meaning of rational, because neither calculation nor deliberative thought is involved, the be-havior of the expert is certainly not irrational. The writings of Schon1983) about knowledge-in-action characterizes the behavior of the expertpractitioner.

Experts do things that usually work, and thus when things are pro-ceeding without a hitch, experts are not solving problems or makingdecisions in the usual sense of those terms. They "go with the Con," asthey say in California. When anomalies occur, when things do not workout as planned or something atypical is noted, deliberate analytic proces-ses are used in the situation. But when things are going smoothly, expertsrarely appear to be reflective about their performance.

Summary. This general stage theory about the development of exper-tise is derived from speculations about how expert systems in the field ofartificial intelligence can be created. Its application to pedagogy is still un-known. Certainly, like all stage theories, the duration of time spent in astage can be expected to vary widely. The more important question iswhether the stages make sense, rather than whether the trip from noviceto expert takes two years or five. In addition, we should note that a personwho is typically at one stage of development may, in particular situations,show characteristics of individuals who are at anorner stage of develop-ment. Also, expertise is considered to be highly contextualized. It may nottransfer from situation to situation very well. This theory is presented be-cause 1 believe it is supported by data my colleagues and I have collectedduring the last few years, and because I believe it has heuristic value forteacher education. Let us examine the supporting data before we discussimplications for policy.

612

Studies of Expertise Many of our studies have used three groups of subjects: Expert, noviceand postulant teachers. Expert mathematics and science teachers at thesecondary level were identified through nomination and chosen afterclasFroom observations were made. Novice teachers were first-yearteach s who had exc silent evaluations when they were student teachers.Postulant teachers were those who worked in industry as engineers, com-puter specialists, and scientists who wanted to teach but were not inter-ested in going though the regular teacher education program. These wereindividuals with subject matter expertise but no pedagogical knowledge ortraining. They are the equivalent to those in many states who enter class-room teaching through an alternative certification route, bypassing col-leges and schools of education. The equivalent stage of development ofthese individuals, in terms of the developmental theory just discussed, isthat of completely unprepared novice. These are the greenest of green,the rawest of raw recruits. They anchor the bottom of a continuum goingfrom ignorant to expert in a particular field. The teachers whom we callnovices in our studies would probably span the categories of novice andadvanced beginner in terms of the theory of development just presented.In a number of are our data support the assertions made in the generaltheory, though we worked without any knowledge of that theory. Let usnow examine the data.

Interpreting classroom phenomena. Because of a lack of experience,the novice is predicted to have trouble in interpreting events. Untilepisodic knowledge is built up and similarities can he recognized acrosscontexts, we can expect confusion to characterize the novices' interpreta-tions of classroom phenomena when compared with the interpretations ofexperts. We obtained data like that in a study where experts, novices, andpostulants watched three television screens showing the same lesson si-multaneously (Sabers, Cushing, & Berliner, 1988). The center screen

I Funding for these studies was provided by the Spencer Foundation, to whom we aregrateful. My colleagues at the University of Arizona included K Carter, P B Clarndge, KCushing, J. George, S. Pinnegar, D. Sabers, & P Stein.

13

`Postulant teachers

. . . with subject

matter expertise but

no ped -logical

knowledge or

training . . . anchor

the bottom of a

continuum going

from ignorant to

expert . .

showed a teacher instructing, taped from the back of the class. Only themiddle group of students was visible. The screen on the right showed onlythe students on the right of the room and the screen on the left showedonly the students on the left of the room. When played simultaneously, acomplex auditory and visual environment was created. Our subjectstalked aloud during part of the time and answered some directed questionsas well. Differences were found among the three groups. For example, thenovices often made contradictory statements about what they had ob-served, especially when they were asked about instructional or manage-ment events within the classroom. The novices experienced difficulty inmaking sense A their classroom observations and in providing plausibleexplanations about what was occurring within the classroom. For ex-ample, when novices were asked to describe the learning environment inthe classroom we heard:

Novice: It looked.. .1 wouldn't call it terribly motivating. It was,well, not bored, but not enthusiastic.

Novice: Very positive as well as relaxed. Very positive. . .it's goodto be able to focus [student] energy into a group situation, yet atthe same time, accomplishing the work that they need to do forthe class and also lending to the relaxed feeling of the classroom.

Such contradictory interpretations were common. Even more dis-crepancy was noted when participants were asked to describe the students'attitudes toward this class. For example:

Novice: It didn't look like it was a favorite etas; for most of them.One boy looked kind of like, "Oh no, it's not this class again."They didn't look overwhelmingly enthusiastic to be there.

Novice: They seemed pretty excited about the class, excited tolearn, and a lot of times it's hard to get students excited aboutscience, but this teacher seems to have them so that they are ex-cited about it. They're willing to work and they want to learn.

As a group, novices seemed unable to make much sense of what theysaw. They experienced difficulty monitoring all three video screens simul-taneously. Thus, they often reported contradictory observations and ap-peared confused about what they were observing and about the interpreta-tions that could be made of their observations.

Postulants are even less familiar with classroom events than novices.Perhaps that is why they appeared even more overwhelmed than thenovices. Many of the postulants expressed difficulty or an inability tomonitor all three video screens simultaneously. Generally, they appearedable to focus on and make sense of only one video screen. Because thislimited their observations, they also made errors and coli.radictory state-ments when they were asked about specific events.

Experts, however, did not demonstrate any confusion or difficulty inmaking sense of the classroom (video) observations. They responded ef-fortlessly and fluidly. They not only made more comments about what washappening (experts averaged 27.4 comments during this segment of thetask; novices, 20.5 comments; postulants. 23.2 comments), but their com-ments were more detailed and descriptive than those of the other twogroups. The experts appeared comfortable both describing what they ob-served and interpreting events in terms of classroom instruction andmanagement. For example, during a second viewing where the expertstalked aloud, one commented:

Expert: Left monitor again. . .1 haven't heard a bell, but the stu-dents are already at their desks and seem to be doing purposefulactivity, and this is about the time that I decided they must be anaccelerated group because they came into the room and startedsomething rather than just sitting down or socializing.

In fact, just as this expert inferred, we had recorded an acceleratedgroup of students. It was a science classroom for students identified asGATE (Gifted and Talented Education) students.

In addition, experts were able to monitor the sounds from both theteacher and students more accurately than did novices or postulants.

9 15

. . . novices

. . . often reported

contradictory obser-

vations and

appeared confused

. . . Experts,

however, . . .

responded effort-

lessly and fluidly . '

Given their ability to monitor three video screens simultaneously for bothvisual and auditory cues, experts seemed less confused than either thenovices or the postulants and thus were better able to interpret classroomevents.

We were particularly impressed with how effortlessly experts could in-terpret and evaluate events and behaviors in real time. For example:

Expert: In the monitor in the middle, it might have been a goodidea to start out class with measuring the height of these plantsthat they're growing while roll is being taken, so that you're notwasting or having a bunch of dead time at the end of the class.

Expert: Again, viewing the middle monitor, I think there is anindication here of the type of structure of this classroom. It's pret-ty loose. The kids come in and go out without checking with theteacher.

Expert: I'm looking at the left monitor...1 think that this is a partof a continuing activity from the day before probably, becausethey know exactly what they're doing without any instructionsfrom the teacher.

Expert: On the left monitor, the students' notetaking indicatesthat they have seen sheets like this and have had presentationslike this before; it's fairly efficient at this point because they'reused to the format they are using.

In contrast, postulants and novices usually gave a step-by-step ac-count of what was happening, as though they were announcing what theywere viewing to someone who could not see the screen. In this respectthey were reminiscent of radio announcers reporting an athletic event.The comments of postulants and novices indicated that perhaps they haddifficulty in interpreting events and behaviors. Their discussions lackedthe inferences, conclusions, evaluations, and suggestions that appearedwith frequency in the protocols of the experts.

10 1 6

Another of our studies also showed this difference in interpretivecompetency between experts and novices (Carter, Cushing, Sabers, Stein,& Berliner, 1988). In this study we had our subjects look at a series ofslides depicting science or mathematics instruction during a class periodin a high school. The subjects held a remote control and were told to viewabout 50 slides at their own pace, stopping to comment on any they foundinteresting. Novices and postulants seemed to show no particular patternin what they stopped to comment on, and they showed the same kinds ofcontradictions in their interpretations that we found in the study usingvideotapes. For example, one novice might say, "Everything looks fine,they're all paying attention," and another novice might say, "It looks likethey're starting to go off task, they're starting to drift." A pattern wasnoted among the experts that was quite different. The experts, more oftenthan the subjects in the other groups, found the same slides worth com-menting about and had the same kinds of comments to make. For ex-ample:

Expert: [Slide 5] It's a good shot of both people being involvedand something happening.

Expert: [Slide 5] Everybody seems to be interested in what they'redoing at their lab stations.

Expert: [Slide 5] Everybody working. A positive environment.

Expert: [Slide 51] More students with their books closed, theirpurses on their desks, hands folded, ready to go.

Expert: [Slide 51] Must be the end of class and everybody is get-ting ready for the bell to ring.

This reduction in variance by the experts is particularly noteworthy.It means they have learned to pay attention to some of the same thingsand to interpret visual stimuli in the same way. This similarity in what isattended to and how it is interpreted is what we want when we visit an ex-pert ophthalmologist or an expert tennis player for advice. Postulants,

novices, advanced beginnersanyone in the early stages of skill acquisi-tionsimply will not have acquired enough experience for that.

Discerning the importance of events. Related to the problem of in-terpreting classroom phenomena is the problem of discerning in complexenvironments what is important to attend to. The theory discussed sug-gests that this is one of the marks of novices, distinguishing them fromthose at higher levels of development . We have evidence for that fromour study of visual processing of classroom information. We showed a slideof a classroom scene very rapidlyfor under 1 secondand asked oursubjects what they saw. The responses of the postulants and novices wereclearly descriptive and usual'', quite accurate.

Postulant: A blond-haired boy at the table, looking at papers.Girl to his left reaching in front of him for something.

Novice: [It's] a classroom. Student with back to camera workingat a table.

Novice: A room full of students sitting at tables.

In contrast to these literal descriptions, typical of novices and pos-tulants, some of our expert teachers gale more than mere descriptions andtheir viewing was more organized.

Expert: It's a hands-on activity of some type. Group work with amale and female of maybe late junior high school age.

Expert: It's a group of students maybe doing small group discus-sion on a project as the seats are not in rows.

For experts, the information that was often deemed important wasthat with instructional significance, such as the age of the students or theteaching/leaming activity in which they were engaged. Blond hair andposture were simply not important. In this sense, our experts were acting

12 18__ _____-_;4

like experts in other fields do, responding not to the literal characteristicsof a situation but only to those aspects that are important.

Experts seemed to have organizing frameworks for viewing. One suchframework was around the concept of work. They commented on "stu-dents working at the blackboard," "students working independently,""teacher looking over a person working in lab," etc. Work appeared to be asalient, organizing concept for the experts as they viewed these slides.

For the most part, postulants' descriptions were characterized bygreater detail about the more static features of the classroom environ-ment. Students were one aspect of the environment, but in contrast to ex-perts, students and their work-related actions were not the prominent fea-tures of their descriptions. In fact, students and their work involvementappeared to be no more salient to postulants than did the physical sur-roundings of the classroom, the equipment, the windows and desks, andthe charts on the wall. Apparently, all the visual stimuli presented hadequal informational value. The postulants were not able to distinguishwhat was important from what was not in the booming, buzzing confusionof life in classrooms. This inability to sort out the forest from the trees is acharacteristic of those in early stages of skill development (Lesgold, 1934).

Using routines. In part, the apparently effortless and fluid perfor-mance of the experts is because of their use of routines that make etass-rooms appear to function smoothly. Leinhardt and Greeno (1986), for ex-ample, in studying elementary school mathematics lessons, compared anexpert's opening homework review with that of a novice. The expertteacher was found to be quite brief, taking about one-third less time thanthe novice. This expert was able to pick up information about attendanceand who did or did not do the homework. She was also able to identifywho would need help later in the lesson. She elicited mostly correctanswers throughout the activity and managed to get all the homeworkcorrected. Moreover, she did so at a brisk pace and never lost any controlof the lesson. She had routines to record attendance, to handle choralresponding during the homework checks, and for hand raising to get at-tention. This expert also used clear signals to start and finish the lesson

13 19

The postulants

were not able to dis-

tinguish what was

important from

what was not in the

booming, buzzing

confusion of life in

classrooms.'

segments. In contrast, when the novice was enacting an opening home-work review as part of a mathematics lesson, she was unable to Jeterminewho did and did not do the homework, had problems with taking atten-dance, and asked ambiguous questions that led to her misunderstandingthe diffir ilty the homework. At one time the novice lost control of thepace. She never learned which students would have difficulty later in thelesson. Of importance is that the novice showed lack of familiarity withwell-practiced routines. She seemed not to have habitual ways to act. Stu-dents, therefore, were unsure of their roles in the class.

In one of our studies (Berliner, 1988), we had experts, novices, andpostulants teach a short lesson on probability to about 15 high school stu-dents. The experts were unhappy about their participation in this task, inpart, because the students they had to teach were not the ones they taughtregularly. Their own students were trained in routines to make the class-room run smoothly. One expert, reflecting (in what was wrong with thetask, said:

Expert 5: My expectations when a kid comes into my classroomfor math is that he has pencil and paper rzady at all times becauseI make them take notes, just as you do in social studies. They havepractice problems. And this is kind of tough 'cause I don't knowwhat was the routine these kids were used to, you know? . .. Youknow, with the kids that are used to your routine, yt can standup and talk for 15 or 20 minutes, and by your questioning techni-ques and by having them work with guided practice at their desks[you keep them working]. But these kids didn't know me, and theydidn't know the way that I operate, that all are supposed to par-ticipate, and why, and that they're all supposed to be on task [con-stantly].

The well-practiced routines of expert surgeons, ice skaters, tennisplayers, and concert pianists (Bloom, 1985), no less than expert teachers,are what give the appearance of fluidity and effortlessness to their perfor-

14 20

mance. What looks to be so easy for the expert and so clumsy for thenovice is the result of thousands of hours of experience and reflection.

Predicting classroom phenomena. The general theory of skill de-velopment also suggests that experience should provide experts morecapability for making better assumptions and hypotheses about classroomphenomena and student behavior, assumptions and hypotheses that arenot obvious to novices. Because txperience leads to recognition of sim-ilarities, the expert learns the probability that certain events or stimuli areassociated with certain other events or stimuli.

We have some data that pertain to this issue from the study where ex-pert, novice, and postulant teachers looked at slides and reported whatthey saw. Experts made many more assumptions, hypotheses, and predic-tions than did the others. They reported, for example:

Expert: Students were not seated in the traditional type of seatingarrangementone that would normally be used for a lecture-typestyle of teaching. So from the seating arrangement, I assumed thatthey must have been involved in some activity other than a tradi-tional type of a lecture.

Expert: . . .there aren't a whole lot of humorous math problems,so I assumed a couple of the students must have been talkingfrom their facial expressionsabout something other than the as-signment.

Expert: I assumed it was the teacher's desk because [it] was faced aiiffirent way from where the students' desks were faced and be-cause of where it was placed in relationship to the chalkboard.

In many other comments made by experts the terms "assume" or"predict" or "hypothesize" could be substituted for the actual words used,as illustrated below:

Expert: It looked like not all students were focused on the same

2115

`What looks to be

so easy for the ex-

pert and so clumsy

for the novice is the

result of thousands

of hours of ex-

perience and

reflection.'

thing, some were faced forward, some were faced back, some werelooking off to the side. So it didn't seem like they were all (so I as-sumed/ hypothesized /predicted they were nod having any sort of groupactivity.

Assuming, hypothesis making, and predicting were much more preva-lent in the language of experts than among novices or postulants. We con-firmed this in another study where expert, novice, and postulant teacherstalked aloud as they thought about how students would answer particularitems from the National Assessment of Educational Progress (Stein, Clar-ridge, & Berliner, 1988). The experts, novices, and postulants differed intheir predictions about the student cognitions used in answering an item.Experts seemed to have a fund of knowledge about the way studentsthought and how those thoughts interacted with the content of thespecific mathematics or science items. In addition, tl..e experts seemedable to think through the misalgorithms that students might apply tosolve a particular problem. The experts had more experience dealing withstudent errors and therefore were able to predict what types of errors stu-dents might make. Novices and postulants rarely discussed the issue ofmisalgorithms that students might apply to solve a problem. The inabilityto predict the kinds of errors that students will make in particular situa-tions is, no doubt, a major deficiency of the beginning teacher. If thetough-to-teach topics for students cannot be distinguished from the easy-to-teach topics and no taxonomy of error types has been built up throughexperience, a teacher is likely to teach in an inappropriate way. Theability to predict how students will think and err is fundamental to a diag-nostic-prescriptive form of teaching. Such knowledge is generally unavail-able to novices.

Judging typical and atypical events. The general theory suggests thatexperts will not pay much attention to things if, in their judgments, theyare going along smoothly. In fact, the expert appears rather unreflective aslong as no problems are perceived. This is another reason that the expertappears to be so effortless at work. Experts apparently pay attention to

16 22

fewer things than do others whose skills are less developed. This does notmean that experts see or hear less, only that they choose to process less ofwhat they encounter. In fact, there is reason to believe that they actuallysee and hear more than do novices (Sabers, Cushing, & Berliner, 1988;Peterson & Comeaux, 1987). In two of our studies we saw some evidenceof this tendency for experts to focus on the atypical rather than the typi-cal.

In one study we asked our subjects to examine extensive informationabout a class that a teacher hac' to leave in the fifth week of a semester(Carter, Sabers, Cushing, Pinnegar, & Berliner, 1987). The simulationwas designed to determine how these expert, novice, and postulantteachers prepared for taking over the class. Experts acted notaNy dif-ferent. While novices and postulants worked diligently to learn all theycould about the various students in the classroom, their achievements,problems, home life, and so forth, the experts invested very little mentaleffort in this activity. They seemed to merge information about studentsinto a "group picture" that they defined as "typical," "normal," or "usual."For example:

Expert: Especially when I start fresh, I start from a clean slate.. .

I like getting a little background on the students in that there aregoing to be severe problems or sameone may need special atten-tion on certain things, you know, learning areas, but, in general,it's a conglomeration of the students. I like to learn from themend develop my own opinions.

Expert: It was a typical classroom, some problem kids that needto be dealt with. And you have to take that into considerationwhen you're developing some kind of plan for them. There are thebright k.ds that were highly self-motivated. There were your shykids. It was a typical class.

Expert: I didn't read the cards. I never do unless there's a com-ment about a physical impairment such as hearing or sight orsomething I get from the nurse. I never want to place a judgment

[u 23

`Experience seems

to change people so

that they literally

"see" differently.'

on the students before they start. I find 1 nave a higher successrate if I don't.

The belief that it is atypicalness that is attended to was supported in adifferent context in the study where slides were used as stimulus materials.Experts tended to describe one of the slides differently from novices orpostulants. For example:

Expert: Not a typical row, not a typical group for conversation...Part of them were facing this way and part of them were facing theother direction, so it wasn't like a typical situation.

Experts apparently have images of how things ought to be. If studentsor classrooms appear to be the way they are supposed to be, they are likelyto be less attended toperhaps even ignored. Experience seems to changepeople so that they literally "see" differently, either by noting atypicalnessquicker (the experts seem to have quicker pattern recognition ability), orby simply not seeing certain ordinary things. Surely that is functional. Inany domain of expertise one must learn through experience, perhapsthousands of hours of experience, what is worth attending to, particularlybecause of the severe biological limits humans have for processing infor-mation.

Evaluating performance: Responsibility and emotions. When thedevelopmental stage of competence is reached, it is said to be accom-panied by a qualitatively different kind of emotionality and sense ofresponsibility for the work of the performer. We found supporting ev-idence, obtained in a curious way, in the study where we had experts,novices, and postulants plan and teach a lesson (Berliner, 1988). The pos-tulants in that study were quite happy about their performance, althoughwe did not rate it highly. Novices were generally affectless in describingtheir experience. They had a task to do and they did it. The experts,however, were quite angry about their participation in the task and quitedisappointed in their performance. We had inadvertently taken away

18

24

some of the experts' edge. First, we had created an artificial teaching situa-tion. Second, according to their standards, they did not have enough timeto prepare the lesson. Third, the students were not trained in the routinesthat make their classrooms hum. One expert expressed his anger by walk-ing away from the study. Another stopped in the middle of the lesson andhad to be coaxed to continue. One started crying during the playback ofthe videotape. All were upset. Two weeks after the study, one expert,when asked what she remembered of her experience, said:

Expect: I just remember it as the worst experience in my entirelife, and I was depressed . . . The things that stick out in my mindare the negative things. I remember just being frustrated thewhole time I taught the lesson ... I don't like what happened. I'vebeen real depressed and down [since then].

Other comments by experts were about their feelings of discomfort,stress, terror, and so forth. In this situation where postulants and noviceswere virtually untouched at any deep emotional level, our experts were af-fected deeply. In addition, they felt that in some way they had let us down.Their sense of responsibility played a part in their feelings as well. Expertteachers, apparently like other experts, show more emotionality abouttheir successes and failures in their work.

Miscellaneous. A listing of the qualitative differences between ex-perts and those whose skills are less developed could be continued forsome time. Some of these differences are related to the increased personalreferences made by the experts, which demonstrate how events from theirpersonal history can be recalled and brought to bear on the problems thatthey face. Some differences are related to the richer, more analytic pro-tocols that the experts provide, as described in a study of experts andnovices conducted by Peterson and Comeaux (1987) and found in someof our studies. Some differences are related to the more principled kinds ofthinking that experts do, also described in Peterson and Comeaux and

19 25

found in our work. Some differences are related to the experts' greaterfacility in understanding students during interactive teaching, because ex-perience apparently has provided them with mental models of the student.as described in the work on teacher-questioning patterns done by Ropoand Kiehela of Finland (1987) and related to findings from our work.

Rather than continue the documentation of differences in the think-ing and performance of individuals at different levels of development, it istime to state the general principle, namely, that important qualitative dif-ferences exist in the thinking and the performance of novices and experts.Research is making clear these qualitative differences. The developmentalprocesses involved in the acquisition of expertise, however, are not yet asclear. The theory of stages in the development of expertise that wasdescribed above is intended to help us think more about that issue. Apoint I have made is that data collected for other purposes were supportiveof that developmental theory. So let us try to use this developmentalmodel and the accompanying set of data to inquire whether interestingpolicy considerations can be formulated.

Policy Considerations Some of the following ideas for policy are tentative, and most need tobe stated with more qualifications than are presented here. Still, theyshould be thought about seriously because a case could be made for each ofthem, given a developmental theory of skill acquisition and certain inter-pretations of the data that is accumulating about the differences betweenexperts and novices.

I. Let us start with a warning about those who come into educationthrough an alternative certification route. These grrenhoms do not pos-sess much pedagogical knowledge. The content knowledge that they pos-sess will not help them much, particularly with hard-to-teach students orin tough teaching assignments. With their minimum of classroom ex-perience, often with no student teaching, and with just a little time spentin observation, they should be considered severely handicapped becauseof their ignorance. Most of the data we have informs us that our pos-tulantsthe rawest recruits to the novice ranklack the perceptual and

20 26

conceptual skills to do well. Policies to support these people after they arein the classroom should be formulated. Additionally, if these kinds ofp,-vle will be relied on to fill shortages in the teaching ranks in the nextfew years, we need short-term training programs of demonstrable effec-:iveness to rapidly compensate for some of their perceptual and concep-tual deficiencies. Such programs ought to be developed at the nationallevel. If the secretary of education and certain state governors want suchpeople in the teaching ranks, then the federal and state governments musteither convince us that the existing information documenting the de-ficiencies of these people is of no consequence and will not prove harmfulto children, or that they can provide training or apprenticeship programsthat overcome the deficiencies these individuals have.

2. Novices are at work to gain experience. Perhaps during studentteaching and during the early months of their first year of classroomteaching we try to present too much knowledge and knowledge that is toosubtle or too sophisticated for them. It is possible that lectures on the waysto adapt instruction to individual differences, readings on how to promotecreativity, and criticisms of performance that focus on the level of mentalprocessing that is communicated in a message by a teacher, are all beyondthe genuine comprehension of the beginning teacher. The point of beginning teaching is the accumulation of experience. That is all beginningteaching is for and that is all we should expect of it. From that experiencecomes the ability to understand what individual differences look and feellike in the classroom, how creative lessons :nteract with other instruction-al goals, and how level of processing can be inferred corn classroom cues.Perhaps the communication of the fine points of eedagogical theory andthe subtleties of pedagogical criticism need to wait until substantial ex-perience has been gained. Early training might better concentrate on (a)perceptual trainingteaching the novice to see what teacher educatorsbelieve is important for later development; and (b) identifying instancesof conceptsteaching the novice to classify things that teacher educatorsconsider important for understanding what occurs when one is a class-room teacher.

2 721

The point of begin-

ning teaching is the

accumulation of

experience.'

There is also some evidence that teachers become expert, in part, bykeeping records of what they did and using these in subsequent years tothink about what to do when teaching something similar. So perhapsbeginning teachers need training and feedback in keeping journals andother records of their teaching behavior and thoughts about teaching.What they probably least need is input from overzealous curriculumreformers and brilliant analysts of teaching who may expect far too muchfrom the beginning teacher.

3. The novices' relative inexperience in a complex environment al-lows a good case to be made for the importance of teaching them standardlesson forms and scripts. The image of the new teacher as the creative les-son planner, eschewing the teachers' manual and bringing a fresh eye tothe creation of curriculum, has already been attacked as unfair by others(Ball & Feiman-Nemser,1986). It is precisely the new teacher who mostneeds the manuals, given with all their deficiencies. Teacher educatorswho would woo the novice away from the inadequate manuals shouldconsider the developmental level of their students and ask where anotherguide can be found for teaching two-column addition with regrouping forthe first time. But having defended the use of the teachers' manuals, wemight even go farther, designing model scripts for beginning teachers touse in teaching the common things that they are most likely to teach.Beginning teachers could gain their experience while under the control ofa script designed by someone who knows better how to teach a particularlesson. Holding an experienced teacher to a scripted lesson, as recom-mended by some behavioral program- concerned about the fidelity of im-plementation, is a terrible idea. This is a policy that will stifle the develop-ment of expertise because such systems require the teacher to always usecontext-free rules. But such systems actually do manage to have teachersdeliver the curriculum that was designed. And for teachers who are intheir first year, it could be a crutch of considerable importance and a wayfor the public to be assured that beginning teachers provide the requiredcurriculum.

2228

4. To do the scripting or provide the practice opportunities necessaryto prepare teachers for their actual performance means knowing what ateacher is going to teach. There is something odd about having to preparea beginning teacher for all six elementary grades or all the mathematicscurriculum in a high school. It is difficult to become an expert withouthaving an area of expertise such as fourth grade, beginning rTading, biol-ogy, or government. We should prepare our novices for particular assign-ments that they will have their first year. Local districts ought to be able tosay to the new teachers they hire, long before the day that school opens,what grade or courses they will need to be prepared to teach. Among thefactors that we are sure distinguished the experts in our studies from otherteachers was their concern for preparation. They were simply eloquentabout the need for time to think through what and how they were goingto teach. They put in many hours every week doing this. But in that mostvulnerable first year, we usually send novices to schools where they do notknow what they will be teaching, have no lesson scripts or lesson pro-totypes to rely on, and are given little planning time during the instruc-tional day. Under these conditions, it will be the rare novice who willhave a successful first year of teaching.

5 . Teacher educators can do more w provide novice teachers withpractice in routines that make classrooms run smoother. Practice inchecking and giving homework assignments, taking attendance, and get-ting students to submit papers efficiently are all relatively easy-to-trainskills. The novice also must learn how to introduce rules for discipline orfor conducting a recitation into the ordinary life of the classroom, andmust practice how to enforce those rules if they are violated. Routinesneed to be learned for handling everyday housekeeping chores and eventsthat have a high likelihood of occurrence, such as a fight or a sudden ill-ness.

Theoretical knowledge is important, but the novice often would bebetter served by practice in routines so that when they run their own classthey can rapidly learn to perform regularly occurring tasks smoothly. Oneimportant function of routines, which every expert has, is that they allow

23 29

`Theoretical

knowledge is impor-

tant, but the novice

often would be

better served

by practice in

routines . . .'

the expert to go into "automatic"to process other information andthink about other aspects of the performance. Experts who haveroutinized whole sequences of behaviorconcert pianists, tennis pros,chess masters, and teachersare free to plan their next steps and monitormore subtle aspects of their performance. Thus, the identification ofteaching routines and an emphasis in training to perform such routines arerecommended.

6. Many of the recent concerns about first-year teachers have resultedin some form of mentor program. The goal has iseen to find expertteachers and pair them in some way with novice teachers in order to helpovercome the problems faced by the novice. We might, however, want tothink again about the qualities that mark an expert, including fluidity inaction, holistic similarity recognition, intuitive and apparently nonreflec-tive patterns of response, and so forth. Perhaps many fine mentor teacherscould be drawn instead from the competent or the proficient groupthose still analytical enough to communicate their reasons for thinkingand acting the way they do.

It has often been noted that many of the great coaches in baseball andfootball were themselves not the most outstanding of athletes. Thegreatest athletes often operate intuitively, making it difficult for any butthe most sensitive and articulate of them to communicate the basis oftheir expertise. (The great Babe Ruth was interviewed in 1928 by CarlSandburg, who asked about the Babe's secret of success. The Babe said "Idon't know, I just pick a good one and sock it.") Competent and proficientathletes have to be more analytical about their performance to achievetheir high status, and it is these who often become the exemplary coaches.A similar phenomenon may hold for teaching as well . Perhaps we shouldinvestigate closely the nature and quality of the mentoring programs wehave and inquire if the proficient and competent are likely to be better atthis task than the expert.

Although an expert teacher may not be an ideal coach, experts can beexcellent models. A classic study of expertise in chicken sexing (Lunn,1948) makes this point abundantly clear. It is virtually impossible to ex-

24 30

plain to another person how to sex a chicken. But if you stand next to anexpert chicken sexer for three months, imitating moves and rememberingdecisions, you will eventually be able to sex 1,000 chickens per hour with98 percent accuracy. Being a good model and being a good coach are dif-ferent characteristics, and we may be confusing them in teacher educa-tion.

7. The recent concerns about teacher evaluation, part of the press foraccountability in our age of educational ferment, have resulted in the useof research-based teacher evaluation instruments. The perfectly logicalbelief undergirding this approach is that because certain teaching prac-tices and methods have been reliably associated with certain desirableeducational outcomes, they should be the focus of evaluation. This stra-tegy may be, in fact, appropriate for the evaluation of novices and ad-vapced beginners, whose job is to learn elements of instruction and whoneed ,:ontext-free rules and procedures to guide them in acquiring ex-perience. But certainly by the time we are dealing with proficient and ex-pc:rt teachers, where context moderates performance and more intuitiveperformance is expected, such evaluation systems are simply irrelevantand an interference. They are not tuned to the developmental needs ofthe teacher.

On the other side of the evaluation spectrum, far from the behavioral,low-inference measurement systems of those who use the research-basedinstruments, are those who argue that connoisseurship and aestheticcriticism are the appropriate evaluation techniques to use with teachers.Here we note that such interpretive evaluation systems may be the onlyway to evaluate those whose competence in the classroom is not in doubt,the proficient and the expert teachers. Because of the level of sophistica-tion these instruments assume about the person being evaluated, such in-terpretive techniques may not be as appropriate for the novice teacher. Atthe very least, feedback from such instruments may be beyond the com-prehension of the typical novice teacher. Advocates of teacher evaluationby means of connoisseurship and criticism, like their more quantitativebrethren, typically have failed to consider the developmental level of the

25 31

`Being a good

model and being a

good coach are

different charac-

teristics, and we

may be confusing

them in teacher

education.'

teacher. The evaluation of teachers is a complex task. As professionalteacher educators. however, we can noticeably improve the process if westop allowing the perpetuation of the myth that a single evaluation instru-ment will suffice for teachers at every level of development.

8. The metaphors for describing teaching now include many morereferences to executives and decision makers, reflecting our increasing un-derstanding of the complexity of the job of the classroom teacher. Thesehave led, naturally enough, to calls for teacher preparation that providesopportunities in decision making. But a developmental model of the ac-quisition of pedagogical skill suggests that such training may be mostrelevant after the level of competence is attained, not before. Decisionmaking, priority setting, and other aspects demonstrating personal controlover the environment are characteristic of the developmental stage ofcompetence. We probably need to think through the scope and sequenceof teacher education experiences in the same way and with the same carethat we develop scope and sequence guides for students from kindergartento twelfth grade. In the process we may learn that certain decision-makinggames and simulations may be more appropriate for the third-year teacherthan the preservice teacher, or that some kinds of case studies to providepractice in analysis and decision making may require knowledge thatpreservice teachers simply do not possess, thus leading to discussions thatare uninformed. Perhaps, also, the practical arguments of the preserviceteacher are not really practical arguments at allmerely prejudices andbeliefs untouched by experience. Thus, developing programs to modifythe practical arguments of teachers may have to wait until some real worldpractice has affected their cognitions.

The question that must be raised while the personnel in preserviceprograms of teacher education struggle to develop reflective practitioners,sensible decision makers, and proficient problem solvers is whether thoseare proper goals for teachers who are more experienced than the novicesin those preservice programs. The research on the development of exper-tise suggests that we have not recognized the limits of the novice and the

26 3 2

potential for growth of the advanced beginner and competent teacher aswe develop teacher education programs.

Conclusion Other policy considerations may be derived from the theory of the ac-quisition of expertise and the supporting empirical data about the dif-ferences between experts and novices. But the point has now been made, Ihope, that developmental differences are real and that they may have im-portant implications for the policies we adopt for the education ofteachers. But a caution is needed as well. Too many educators, upon read-ing Piaget, thought that we should not bother with asking children so-phisticated scientific questions because they did not have the requisite ex-perience to handle them. Fortunately, others recognized that a develop-mental theory does not mean that certain experiences should be avoidedcompletely or that individuals should be exposed to problems only at theappropriate level of their development. Thus, I am not advocating a nar-row form of job training as the curriculum for preservice teacher prepara-tion. I am suggesting that our extensive knowledge base about teachingand teachers be thought of as more or less appropriate to people in dif-ferent stages of their development. I am also suggesting that preserviceeducation may not be the most appropriate place to teach some things,and therefore we rani have to extend our programs of teacher educationfor some time after our students have entered practice. I am suggesting aswell that the forms of evaluation for experienced and beginning teachersmay have to differ. And I am suggesting that experts, revered as they iaaybe, may not always make the best teachers of novices. I am arguing thatthe development of competence out of ignorance and expertise out ofcompetence may take a long time in a profession as complicated as teach-ing. We may be unable to shorten the trip very much because extensiveexperience is fundamental to development, but we certainly ought to helpnurture those willing to undertake the journey by providing training andevaluation appropriate for their level of development.

t

2733

' . . . developmental

differences are real

and . . . they may

have important

implications for the

policies we adopt

for the education of

teachers.'

References Ball, D. L., & Feiman-Nemser, S. (1986, Am!). Using textbooks and teachers' guides: Whatbegmtung elementary teachers learn and what they need to know. Paper presented at themeeting of the American Educational Research Association, San Francisco.

Berliner, D. C. (1986). In pursuit of the expert pedagogue. Educational Researcher, IS , 5-13.Berliner, D. C. (1988, April). Memory for teaching as a function of expertise. Paper presented

at the meeting of the American Educational Research Association, New Orleans.Benner, P. (1984). From novice to expert. Reading, MA: Addison-Wesley.Bloom, B. S. (Ed.).(1985). Developmg talent m young peoge. New York: Ballennne.Carter, K., Cushing, K., Sabers, D., Stein, P., & Berliner, D. C. (1988). Expert-novice dif-

ferences in perceiving and processing visual classroom information. Journal of TeacherEducanon,39 (3), 25-31.

Carter, K., Sabers, D., Cushing, K., Pinnegar, P. , & Berliner, D. C. (1987). Processing andusing informant,- about students: A study of expert, novice and postulant teachers.Teachmg and Teuctier Education, 3 , 147-157.

C111, M. T. H., Glaser, R.,& Farr, M. (Eds.). (in press). The nature of expertise Hillsdale, NJ:Erlbaum.

Dreyfus, H. L., & Dreyfus, S. E. (1986). Mind over machine. New York: Free Press.Leinhardt, G., & Greeno, J. (1986). The cognitive skill of teaching. Journal of Educational

Psychology, 78, 75-95.Lesgold, A. M. (1984). Acquiring expertise. In J. R. Anderson & S. M. Kosslyn (Eds.),

Tutorials in learning and memory (pp. 31-60). San Francisco: W. H. Freeman.Lunn, J. H. (1948). Chick sexing. American Scientist, 36, 280-287.Peterson, P. L., & Comeaux, M. A. (1987). Teachers' schemata for classroom events: The

mental scaffolding of teachers thinking during classroom instruction. Teaching andTeacher Education, 3, 319-331.

Rope, E., & Kiehela, H. (1987, September). Teachers' Questions: Some differences betweenexpert and novice teachers. Paper presented at the meetling of the European Associationfor Research in Learning and Instruction, Teubingen, Germany.

Sabers, D., Cushing, K., & Berliner, D. C. (1988 ). Differences among expert, novice, and pos-tulant teachers in a task characterized by simultaneity, muluthmensionahty and immediacy.Manuscript submitted for publication.

Schon, D. (1983). The reflective practitioner. New York: Basic Books.Stein, P., Clamdge, P. B., & Berliner, D. C. (1988). Teacher estimanon of student knowledge

Accuracy, content and process Manuscript submitted for publication.

3428

";C

15,A

k5:A

:F'"

5'55,4 (1'1.*45;

'.404.;:5'i,

,74

'I

4'

'

,4'

';'

'',"'1,"

,

5t'

'1

.V.;

4

r''':1`.:4d.'`t

.".5';11.1/4

'

,

,,

''f,"'

;1'

1''',

',

-to'',,;'-',

'i'

''-''

"';$:,!; '.'..i''

,,

,.

,

.'0,''' ; ''.'

*',

,,

t-

,s,

,"41)

'ff'k

t'

'